DraftDraft 1 Corticosteroid-binding globulin levels in North American sciurids: implications for the flying squirrel stress axis Lanna M. Desantis1*, Jeff Bowman2, Erin Faught3, Rudy

Draft

Corticosteroid-binding globulin levels in North American

sciurids: implications for the flying squirrel stress axis

Journal: Canadian Journal of Zoology

Manuscript ID cjz-2017-0300.R1

Manuscript Type: Article

Date Submitted by the Author: 05-Feb-2018

Complete List of Authors: Desantis, Lanna; Trent University, Environmental & Life Sciences Graduate Program Bowman, Jeff; Ontario Ministry of Natural Resources and Forestry, Wildlife Research and Development Section Faught, Erin; University of Calgary, Department of Biological Sciences Boonstra, R.; University of Toronto Scarborough, Centre for the

Neurobiology of Stress, Department of Biological Sciences Vijayan, Mathilakath; University of Calgary, Department of Biological Sciences Burness, Gary; Trent University, Department of Biology

Keyword: Glaucomys sabrinus, Glaucomys volans, glucocorticoids, northern flying squirrel, physiological ecology, southern flying squirrel, western blot

https://mc06.manuscriptcentral.com/cjz-pubs

Canadian Journal of Zoology

Draft

1

Corticosteroid-binding globulin levels in North American sciurids:

implications for the flying squirrel stress axis

Lanna M. Desantis1*

, Jeff Bowman2, Erin Faught

3, Rudy Boonstra

4, Mathilakath

M. Vijayan3 and Gary Burness

5

1Environmental & Life Sciences Graduate Program, Trent University, Peterborough, ON K9L

0G2, Canada

2Ontario Ministry of Natural Resources and Forestry, Trent University, DNA Building,

Peterborough, ON K9L 1Z8, Canada

3Department of Biological Sciences, University of Calgary, Calgary, AB T2N 1N4, Canada

4Centre for the Neurobiology of Stress, Department of Biological Sciences, University of

Toronto Scarborough, Toronto, ON M1C 1A4, Canada

5Department of Biology, Trent University, Peterborough, ON K9L 0G2, Canada

* Corresponding Author

E-mail: [email protected]

Phone: 705-748-1011 ex. 7288

Page 1 of 26



Draft

2

Corticosteroid-binding globulin levels in North American sciurids:

implications for the flying squirrel stress axis

Lanna M. Desantis, Jeff Bowman, Erin Faught, Rudy Boonstra, Mathilakath M. Vijayan and

Gary Burness

Abstract: Corticosteroid-binding globulin (CBG) helps to regulate tissue bioavailability of

circulating glucocorticoids (GCs), and in most vertebrates, ≥ 80-90% of GCs bind to this protein.

New World flying squirrels have higher plasma total cortisol levels (the primary corticosteroid in

sciurids) than most vertebrates. Recent research suggests that flying squirrels have either low

amounts of CBG or CBG molecules that have a low binding affinity for cortisol, since this taxon

appears to exhibit very low proportions of cortisol bound to CBG. To test whether CBG levels

have been adjusted over evolutionary time, we assessed the levels of this protein in the plasma of

northern (Glaucomys sabrinus Shaw, 1801) and southern (G. volans L., 1758) flying squirrels

using immunoblotting, and compared the relative levels among three phylogenetically related

species of sciurids. We also compared the pattern of CBG levels with cortisol levels for the same

individuals. Flying squirrels had higher cortisol levels than the other species, but similar levels of

CBG to their closest relatives (tree squirrels). We conclude that CBG levels in flying squirrels

have not been adjusted over evolutionary time, and thus, the uncoupling of CBG levels from

cortisol concentrations may represent an evolutionary modification in the lineage leading to New

World flying squirrels.

Keywords: Glaucomys sabrinus, Glaucomys volans, glucocorticoids, northern flying squirrel,

physiological ecology, Sciuridae, southern flying squirrel, western blot

Page 2 of 26



Draft

3

Introduction

The stress axis (or the hypothalamic-pituitary-adrenal [HPA] axis) allows vertebrates to

survive and reproduce by providing a physiological mechanism through which they can react to

stressors in their environment. This axis sustains the initial “fight or flight” response initiated by

the sympathetic nervous system when a stressor is perceived, by releasing glucocorticoids (GCs;

the stress hormones cortisol and/or corticosterone). GCs mobilize and trigger the replenishment

of depleted energy stores to allow the organism to deal with the stressor and restore homeostasis

(Sapolsky et al. 2000).

To avoid long-term negative effects, GC levels are regulated in two ways. The first is via

the negative feedback mechanism of the stress axis, in which the production and release of GCs

is inhibited once the stressor subsides (Sapolsky et al. 2000). The second is via a carrier protein,

corticosteroid-binding globulin (CBG), that is produced in the liver and circulates in the blood,

binding GCs with high affinity (Westphal 1983). The binding of GCs to CBG limits the passage

of GCs through cell membranes, thereby increasing its half-life in circulation and also restricting

its biological action (Mendel 1992; Perogamvros et al. 2012; Breuner et al. 2013; reviewed by

Hammond 2016). In the current study, we focus on this second mode of regulation.

Most vertebrates produce sufficient CBG to bind ≥ 80-90% of their basal circulating GC

levels, thus protecting their tissues from these hormones when the body does not require their

use. The relationship between GCs and CBG is evolutionarily highly conserved (Desantis et al.

2013). An exception to this “90% bound” rule is a number of New World monkeys (5 species in

4 genera, in 3 families) that have cortisol levels higher than in the majority of vertebrates studied

(Chrousos et al. 1982), yet proportions bound to CBG being ≤ 5% (Klosterman et al. 1986). CBG

in these monkeys has an exceptionally low affinity for cortisol (Robinson et al. 1985;

Page 3 of 26



Draft

4

Klosterman et al. 1986; Rosner et al. 1986), presumably due to mutations in the CBG molecule

(Robinson et al. 1985; Hammond et al. 1994). Despite having very limited buffering capacity

against the biological effects of circulating cortisol, a much-reduced affinity of the tissue

receptors for cortisol provides New World monkeys with the needed protection in lieu of CBG

binding (Scammell et al. 2001). However, the most parsimonious explanation for their current

physiology is that high circulating levels of cortisol evolved as a response to the decreased

sensitivity of their receptors to the hormone through evolutionary processes (discussed by

Desantis et al. 2013).

New World flying squirrels, the northern (Glaucomys sabrinus Shaw, 1801) and southern

(G. volans L., 1758) flying squirrel, represent a second lineage in which only about 5-10% of

cortisol is bound to CBG, suggesting their tissues may be exposed to high levels of

corticosteroids on a regular basis (Desantis et al. 2013). Mechanistically, it is unclear whether

low binding of cortisol to CBG in flying squirrels is due to low circulating CBG levels or to

CBG molecules with a low affinity for cortisol, and thus far, CBG has only been quantified in

flying squirrels using an assay which relies upon labeled cortisol binding tightly to CBG (i.e.

with high affinity) for quantification of the protein (Desantis et al. 2013).

In the present study, we sought to clarify whether low binding of cortisol to CBG in

flying squirrels is due to low levels of CBG or to the low affinity of the protein for this steroid.

We examined plasma CBG levels in North American flying squirrels and compared these levels

with those of three other sciurid (squirrel) species: two are closely related to flying squirrels - the

North American tree squirrels (the red squirrel, Tamiasciurus hudsonicus Erxleben, 1777, and

the eastern grey squirrel, Sciurus carolinensis Gmelin, 1788); and one that is distantly related

and used as an evolutionary outgroup (the eastern chipmunk, Tamias striatus L., 1758; Fig. 1).

Page 4 of 26



Draft

5

We also measured plasma total cortisol concentrations from these same individuals, allowing us

to qualitatively assess the relative levels of CBG and cortisol across species.

We test three competing hypotheses. First, flying squirrels do not produce CBG, likely as

a result of a gene deletion. If this were true, we would not expect to observe plasma CBG

expression for flying squirrels, and thus, the 5-10% binding found by Desantis et al. (2013) was

likely a result of cortisol binding nonspecifically to some other protein (e.g., serum albumin).

Second, flying squirrels do produce CBG and the relationship between CBG levels and

corticosteroid concentrations is evolutionarily conserved such that their physiology exhibits an

80-90% bound scenario. In other words, species with higher basal corticosteroid concentrations

will have correspondingly higher CBG levels. If this were true, then we predict that the

relationship between cortisol and CBG would be the same across the five sciurid species. Third,

flying squirrels produce CBG and the relationship between CBG levels and cortisol

concentrations is evolutionarily derived. Under this scenario, we predict that CBG levels would

be lower-than-expected given their cortisol concentrations, when compared with

phylogenetically-related species.

Materials and methods

Study Sites

The principal study site was located near Mississagua Lake in the Kawartha Lakes

Region of south-central Ontario, Canada (44º41’18”N 78º20’8”W), and was used to capture

northern and southern flying squirrels, red squirrels and chipmunks. Live trapping was conducted

in a portion of contiguous forest just west of Kawartha Highlands Provincial Park on the

southern edge of the Canadian Shield (the transition zone between the Carolinian forests of

Page 5 of 26



Draft

6

southern Ontario and the Boreal forests of northern Ontario). Grey squirrels were live-captured

in a patch of mixed deciduous forest on the endowment lands of the University of Toronto

Scarborough, in Scarborough, Ontario, Canada (43º47’05”N / 79º11’18”W).

Live-trapping and Blood Sampling

Blood plasma samples were collected during the fall of 2008 and 2009, between mid-

September and mid-November. To minimize variation in cortisol and CBG levels that may arise

among individuals and between seasons and sexes, all animals used in our analyses were

sampled in the non-breeding season. All individuals were adult males, except for two adult

female grey squirrels because of a lack of available male plasma for this species. We chose to

retain the 2 females in our analysis, as the variation among the four individual grey squirrels (2

males and 2 females) was minimal (mean cortisol ± SE = 59.34 ± 2.26) ng/ml; mean CBG levels

= 664.68 ± 62.12 arbitrary units), when compared with intra-specific variation in the other four

species. Samples sizes were 3 southern flying squirrels, 3 red squirrels, 4 northern flying

squirrels, 4 eastern chipmunks, and 2 male and 2 female eastern grey squirrels.

Tomahawk live-traps (Model 102, Tomahawk Live Trap Company, Tomahawk, WI,

USA) were used to capture all species and were baited with peanut butter and whole peanuts.

Longworth live-traps were also used for chipmunks with the same bait. Traps were set just prior

to dawn for diurnal species (red and grey squirrels, and chipmunks), or dusk for nocturnal

species (northern and southern flying squirrels) and checked 1-2 hours after first light or after

dark, respectively. For red squirrels and chipmunks, two traps of each type per location were

placed on the ground along a trap line adjacent to a sand road used for cottage access. Each of

the 17 locations was approximately 100-150 m apart (depending on the terrain). Live-traps for

Page 6 of 26



Draft

7

flying squirrels were fastened with bungee cords to wooden platforms that were mounted on tree

trunks approximately 2 m above the ground. A portion of the same trap line described above

was used for flying squirrel capture (10 locations with two traps at each), as well as a trapping

grid located nearby, with 68 traps spaced 30 m apart (one trap per grid point). For grey squirrels,

twelve traps were placed singly on the ground throughout the forest patch approximately 10-20

m apart and where the substrate provided a protected place to set a trap.

Each day or evening, all captured animals were transported to a central location near the

sampling grid or trap line for processing. The time for transport of the animals to this location (~

15 min), plus the time these individuals were allowed to sit undisturbed in their traps before

blood sampling began (~30 to 45 min), provided a standardized acclimation period across all

sampling days. Thus, while blood samples were admittedly from animals that experienced

capture stress, there was a consistent period of time after transport and prior to blood sampling,

which resulted in minimal individual variation in cortisol levels (Fig. 2). This approach has been

used previously in other comparative studies of mammalian stress physiology (e.g., Delehanty

and Boonstra 2009). Therefore, including the time animals were in traps prior to traps being

checked, the total time spent in captivity prior to blood sampling was 1-3 hours, and this range

was identical across all species. On average, 3-6 individuals were captured per day/night.

Squirrels were anesthetized with isoflurane (Abbott Laboratories, Montreal, QC, Canada), and

blood was then drawn from the sub-orbital sinus (200–500 µL) with heparinized Pasteur pipettes

and kept on ice until sampling was complete (~ 2-4 hours). Plasma was removed after

centrifugation, and samples were stored at -80 °C until processing.

Research on live animals followed the guidelines of the the Canadian Council on Animal

Care, and was approved under a University of Toronto Animal Use Protocol (# 20007021).

Page 7 of 26



Draft

8

Measurement of Plasma Total Cortisol

Total plasma cortisol was measured using a commercially available radioimmunoassay

(Clinical Assays GammaCoat Cortisol 125

I RIA Kit; DiaSorin, Stillwater, MN, USA). This kit

was validated for parallelism with plasma from all five species. Tests for differences between

slopes on log-transformed data showed that serially diluted plasma curves for all species were

parallel to the assay standard curve (southern flying squirrels: F1,11 = 0.40, P = 0.54; northern

flying squirrels: F1,11 = 0.75, P = 0.41; red squirrels: F1,13 = 1.68, P = 0.22; grey squirrels: F1,13 =

0.17, P = 0.68; chipmunks: F1,13 = 0.32, P = 0.58). The intra- and inter-assay coefficients of

variation (CV) were 5.3% and 10.3%, respectively.

Plasma Corticosteroid-binding Globulin Levels

We used an affinity-purified polyclonal antibody generated from the amino acid sequence

for grizzly bear (Ursus arctos L., 1758) CBG (gbCBG antibody) that was synthesized and

validated by Chow et al. (2010). The cross reactivity of the gbCBG antibody with squirrel

plasma was confirmed by western blotting (described below) with serially diluted plasma

samples for all five species (5, 10, 20 and 40 µg total plasma protein).

Total plasma protein concentrations of unknown samples were determined by the

bicinchoninic acid (BCA) method using bovine serum albumin (BSA) as the standard, and

plasma proteins separated by SDS-PAGE. Briefly, plasma samples were prepared in 2×

Laemmli’s buffer with ß-mercaptoethanol (5%), and boiled at 95ºC for 10 min. Samples were

loaded onto 8% reducing polyacrylamide gels according to established protocols (Boone and

Vijayan 2002). A low-range molecular weight marker (BLUeye Prestained Protein Ladder for

Page 8 of 26



Draft

9

Tris-Glycine buffer, GeneDirex) was also loaded to confirm the molecular mass (kDa) of the

protein detected. Unknown samples were loaded at 20 µg total protein, and the same positive

control (an individual southern flying squirrel previously used to confirm cross reactivity of

squirrel plasma with the gbCBG antibody) was loaded at 40 µg total protein on every gel to

normalize for inter-gel variability. Plasma proteins were separated (200 V for 35 min; Mini

Protean III [Bio-Rad]) using a discontinuous buffer. The separated proteins were transferred to a

0.45 µm pore size nitrocellulose membrane (Bio-Rad) using a Transblot SD Semi-Dry

Electrophoretic Transfer Cell (Bio-Rad) and transfer buffer (25 mM Tris [pH 8.3], 192 mM

glycine, 10% v/v methanol). Transfer efficiency and equal loading of protein were confirmed by

Coomassie brilliant blue (Bio-Rad) staining of the polyacrylamide gel, and Ponceau S (Bio-Rad)

staining of the nitrocellulose membrane.

Membranes were rinsed and blocked with 5% skim milk in TTBS (20 mM Tris, pH 7.5

[Fisher], 300 mM NaCl [Sigma], 0.1% (v/v) Tween 20 [Bio-Rad]). Blots were probed with a

polyclonal rabbit anti-grizzly bear (gb) CBG (primary antibody, diluted 1:500) for 1.5 h at room

temperature. The blots were washed with TTBS (3x 10 min) and incubated with goat anti-rabbit

IgG conjugated horseradish peroxidase (HRP; secondary antibody, diluted 1:3000; Bio-Rad) for

1 h at room temperature. Blots were further washed (2x 10 min in TTBS, and 1x 10 min in TBS)

and the proteins detected using Clarity™ Western ECL Substrate blotting detection reagent (Bio-

Rad). Protein bands were scanned using a chemiluminescence imager (G:Box Chemi XX6,

Syngene) and quantified using the image editor ImageJ 1.49v for Mac (Rasband 2015). The band

densities were normalized to the same reference plasma CBG level (adult male southern flying

squirrel) used on each gel, and the values are displayed as arbitrary units.

Statistical Analyses

Page 9 of 26



Draft

10

To distinguish between hypotheses 2 and 3, we used both ANOVA and regression

analysis. First we assessed whether species differed in plasma total cortisol and CBG levels

using one-way ANOVAs. If flying squirrels have higher cortisol concentrations and CBG levels

than the other species, this would support our second hypothesis. Conversely, if flying squirrels

have higher cortisol, but their CBG levels are not higher than the other species, this would

support our third hypothesis. We tested for normality using the Kolmogorov-Smirnov normality

test, and all data were normally distributed (all p > 0.10). We tested for equal variance using

Levene’s test. All species groups had equal variance for total cortisol (F4,14 = 0.37, p = 0.82), but

not for CBG (F4,13 = 4.25, p = 0.02). We thus used Welch’s Test (the Welch ANOVA F-test for

unequal variances) to analyze the CBG data. The Tukey-Kramer HSD Multiple Comparison Test

was used for post-hoc analysis.

As a second method to distinguish between hypotheses 2 and 3, we assessed whether the

five species showed a similar relationship between CBG and cortisol. We generated a scatterplot

of CBG vs. total cortisol levels for individuals of all species, and performed a linear regression

among the three non-flying squirrel species (red squirrels, grey squirrels and chipmunks).

Because the three comparative species are considered to have a typical relationship between

CBG and GCs (Desantis et al. 2013), we used this regression line to qualitatively evaluate

whether the distribution of data points for the two flying squirrel species followed the same

relationship trend. We were not able to rigorously account for the phylogenetic non-

independence of the data with only five species-level data points, but our qualitative display of

the relationship helped to support the results of the ANOVA analysis described in the previous

paragraph. All tests were performed using JMP version 12.1.0 (2015, SAS Institute Inc., Cary,

NC, USA). Significance was set at p < 0.05.

Page 10 of 26



Draft

11

Results

Total cortisol levels in plasma varied among the five species (F4,13 = 99.38, p < 0.0001),

with southern flying squirrels having the highest cortisol levels (1.7× higher than northerns, 2.4×

reds, 3.0× greys, and 5.7× chipmunks). The most closely related species to the flying squirrels,

red and grey squirrels, had similar levels to one another, but these levels were lower than in both

flying squirrels, and higher than cortisol levels in our evolutionary outgroup, the eastern

chipmunk (Fig. 2; all p < 0.02).

CBG was detectable in all five species (Fig. 3), as indicated by visible bands at 55 kDa.

CBG levels varied among species (Welch’s Test: F4,5.4 = 27.86, p = 0.0009), although the

species-level patterns differed from those of total cortisol. Southern and northern flying squirrels,

and both red and grey squirrels all had similar relative CBG levels (Tukey’s HSD, ns). However,

relative levels of CBG in eastern chipmunks was 5.0× lower than that in southern flying squirrels

(p = 0.0014; Fig. 3), 4.3× lower than in northern flying squirrels (p = 0.0043), and 4.3× lower

than in red squirrels (p = 0.0071). Thus, both flying squirrel species had CBG levels that were

similar to the closely related tree squirrels, even though their cortisol levels were higher than all.

From our regression analysis (Fig. 4), both cortisol and the relative availability of CBG

increased across the three comparative species. Data points for both flying squirrel species fell

below the line; although northern flying squirrels fell much closer to the regression line than did

southern flying squirrels. Thus, flying squirrels did not appear to show the same relationship

between the two variables of interest as the other three species, such that their relative levels of

CBG were not proportional to their high cortisol levels.

Page 11 of 26



Draft

12

Discussion

We proposed three competing hypotheses to assess alternative explanations for the

apparently low levels of binding between cortisol and CBG in flying squirrels. Based on our

immunodetection, it was clear that both southern and northern flying squirrels express plasma

CBG (Fig. 3), and thus we rejected the first hypothesis that there was a gene deletion for the

protein in these species. The flying squirrels had plasma CBG levels similar to that of the three

comparative species (Fig. 2). Also, the molecular mass (~ 55 kDa) was similar to that of other

vertebrates, supporting a highly conserved protein (Chow et al. 2010; Moisan 2010). However,

we might have predicted alterations in molecular mass for CBG in flying squirrels. For example,

the New World monkey species that show a similar stress physiology to flying squirrels (high

cortisol and low CBG with immeasurable or very high Kd’s [high values indicative of low

binding]; Robinson et al. 1985; Klosterman et al. 1986; reviewed by Desantis et al. 2013) have

structurally altered CBG molecules that circulate as dimers and thus have higher molecular mass

(Hammond et al. 1994). Our use of denaturing gels in our immunoblotting process did not allow

us to detect the possibility of dimers, but future analysis with native gels might reveal such

structural alterations in flying squirrels. Also, since Desantis et al. (2013) found evidence of

weak or dysfunctional binding of cortisol to the CBG protein in flying squirrels, just as in the

New World monkeys, this suggests the possibility of parallel evolution. Further evaluation of

molecular charge and the 3D structure of the protein might reveal mutations in the tertiary

structure or the ligand-binding site of CBG in flying squirrels, but this was beyond the scope of

our study.

In our five-species comparison, CBG levels increased in parallel with circulating cortisol

concentrations in red squirrels, grey squirrels, and chipmunks. This was expected as the majority

Page 12 of 26



Draft

13

of vertebrates have enough CBG to bind ≥ 80-90% of GCs, and this relationship presumably

represents an evolutionarily conserved strategy (Desantis et al. 2013; Fig. 2). In contrast, flying

squirrel CBG levels were lower relative to cortisol levels (Fig. 2). For example, whereas cortisol

concentrations in southern flying squirrels were 1.7 to 5.7× higher than in the other four species,

their CBG levels was statistically the same as three of them (Fig. 2). Furthermore, when we

compared relative levels of CBG with cortisol (as an index of “percent bound”; Fig. 4), all of the

data points for the flying squirrels fell below the predicted regression line. As a result, CBG in

flying squirrels did not appear to be expressed in proportion to circulating cortisol levels, as seen

in the other three species. This finding supports the conclusion by Desantis et al. (2013) and our

third hypothesis, that low proportions of bound cortisol arose in the lineage that gave rise to New

World flying squirrels, and likely represents an evolutionarily derived state for this relationship.

As cortisol levels increased in northern and southern flying squirrels compared with tree and

ground squirrels, plasma CBG levels remained constant, never evolving to catch up

proportionally to their circulating cortisol concentrations. We assume here, that our inference

about cross-species cortisol levels was not affected by our trapping methods, which required that

we sampled stressed levels.

The suggestion that flying squirrels do not exhibit the typical 80-90% bound relationship

between cortisol and CBG suggests evolutionary divergence in their stress axis, especially with

respect to target tissue responsiveness to corticosteroid action. This is particularly the case within

southern flying squirrels, whose CBG levels, as predicted from cortisol levels of related species,

should have been considerably higher than what we detected. Northern flying squirrels appear to

have higher proportions bound than reported previously (Desantis et al. 2013), which also gives

further corroboration for a dysfunctional binding site proposed by Desantis et al. (2013). Given

Page 13 of 26



Draft

14

that both flying squirrel species appear to have reduced binding capacity for cortisol (Desantis et

al. 2013), one possibility is that the divergence in this trait (i.e., having enough CBG for an 80-

90% bound scenario) began with a mutation in the CBG molecule in the northern flying squirrel,

and was subsequently modified in southern flying squirrels following speciation (northern and

southern flying squirrels are sister species: Arbogast 1999; Kerhoulas and Arbogast 2010). We

also suggest the following mechanistic explanation for this divergence in the stress axis of flying

squirrels. If these species are glucocorticoid resistant (i.e. they require high baseline cortisol

(Desantis et al. 2016) because target tissue receptors are more resistant to the hormone in that

they less readily bind cortisol for biological action), then their high cortisol levels may not

stimulate the synthesis of proportional amounts of CBG (e.g. Smith and Hammond 1992). A

direct measurement of the amount of circulating free cortisol in New World flying squirrel

plasma in future studies would help to confirm or refute the relative levels of plasma CBG

presented here (Hammond 2016).

The current study strongly supports previous evidence that the stress physiology of New

World flying squirrels has undergone evolutionary divergence. Although we cannot directly

compare the relative levels of CBG determined by immunoblotting with cortisol concentrations -

given that one is semi-quantitative - we are confident that the pattern shown for the relationship

between these two variables (Figs. 2, 4) is representative of the relative proportions bound for

each species. For example, we know that red and grey squirrels and chipmunks have a binding

capacity of 80-90% (Desantis et al. 2013), and thus have CBG with a high affinity for cortisol

(i.e. a functional binding protein).

There are some caveats to our conclusions. We recognize our sample sizes are relatively

small; however, our estimates of circulating cortisol levels are similar to those reported

Page 14 of 26



Draft

15

previously using similar trapping techniques with larger sample sizes (Boonstra and McColl

2000; Desantis et al. 2013, 2016; L. Desantis and R. Boonstra, unpublished data). We also

realize that our inference is limited about the relationship between cortisol and CBG levels

because of the differing units of the two variables, with western blot results being semi-

quantitative. Nonetheless, we are confident that our overlay of the two variables (Fig. 2) shows

species-level patterns that are biologically meaningful. This is in part because these patterns are

similar to those reported previously for chipmunks, red and grey squirrels, where CBG levels

were estimated via binding to cortisol (Desantis et al. 2013). Finally, as stated earlier, we assume

that our capture methods have allowed accurate estimation of cross-species trends in stressed

cortisol levels.

The question that remains is just how functional the CBG molecule is in New World

flying squirrels compared with other species. Perhaps there is just enough effective binding of

cortisol to CBG for scenarios where the bound complex is required (e.g. delivery to sites of

inflammation, or for binding to extracellular membrane receptors to allow intracellular delivery

of cortisol; Pemberton et al. 1988; Lin et al. 2010; Perogamvros et al. 2011). Determining

expression levels and the affinity of glucocorticoid and mineralocorticoid receptors (GRs and

MRs, respectively) in target tissues of flying squirrels will help us to understand how they might

cope with an ineffective CBG molecule to act as a buffer. Our prediction is that their tissue GRs

and MRs are altered or down-regulated such that they do not bind cortisol as readily as in other

species, and this helps to regulate the biological function of cortisol despite high circulating

levels of this steroid, just as is the case for New World monkeys (Scammell et al. 2001). Another

possible avenue of exploration is to test the activity of biotransformation enzymes, especially

11ß-HSD1 and 11ß-HSD2, in flying squirrels to determine if inactivation of cortisol is a strategy

Page 15 of 26



Draft

16

adopted by these species to regulate biological activity at the receptors. Knowledge of the

mechanisms that New World flying squirrels use to offset the catabolic effects of cortisol will

provide insight into possible evolutionary limits of plasticity within the vertebrate stress axis.

Acknowledgements

We thank Dr. JK Thomas for assistance with western blots in the laboratory, Dr. TJ

Hossie for comments on phylogenetic independence and analysis, G Keresztesi and J Middleton

for assistance in the field, and the James McLean Oliver Ecological Centre for their hospitality

while collecting samples. This research was supported by the Natural Sciences and Engineering

Research Council of Canada (NSERC; GB, JB, RB and MMV), the Canadian Foundation for

Innovation and Ontario Innovation Trust (GB, RB and MMV), scholarships from the

Government of Ontario (LMD) and NSERC (EF), and the Wildlife Research and Monitoring

Section of the Ontario Ministry of Natural Resources and Forestry (JB).

References

Arbogast, B.S. 1999. Mitochondrial DNA phylogeography of the New World flying squirrels

(Glaucomys): Implications for Pleistocene biogeography. J. Mammal. 80: 142-155.

Boone, A.N., and Vijayan M.M. 2002. Constitutive heat shock protein 70 (HSC70) expression in

rainbow trout hepatocytes: effect of heat shock and heavy metal exposure. Comp.

Biochem. Physiol. C Toxicol. Pharmacol. 132: 223–233.

Boonstra, R., and McColl C.J. 2000. Contrasting stress response of male Arctic ground squirrels

Page 16 of 26



Draft

17

and red squirrels. J. Exp. Zool. 286: 390-404.

Breuner, C.W., Delehanty, B., and Boonstra, R. 2013. Evaluating stress in natural populations of

vertebrates: total CORT is not good enough. Funct. Ecol. 27: 24-36.

Chrousos, G.P., Renquist, D., Brandon, D., Eil, C., Pugeat, M., Vigersk, R., Cutler Jr., G.B.,

Loriaux, D.L., and Lipsett, M.B. 1982. Glucocorticoid hormone resistance during primate

evolution: receptor-mediated mechanisms. Proc. Natl. Acad. Sci. U.S.A. 79: 2036–2040.

Chow, B.A., Hamilton, J., Alsop, D., Cattet, M.R.L., Stenhouse, G., and Vijayan, M.M. 2010.

Grizzly bear corticosteroid binding globulin: cloning and serum protein expression. Gen.

Comp. Endocrinol. 167: 317-325.

Delehanty, B., and Boonstra, R. 2009. Impact of live trapping on stress profiles of Richardson’s

ground squirrel (Spermophilus richardsonii). Gen. Comp. Endocrinol. 160: 176-182.

Desantis, L.M., Delehanty, B., Weir, J.T., and Boonstra, R. 2013. Mediating free glucocorticoid

levels in the blood of vertebrates: are corticosteroid-binding proteins always necessary?

Funct. Ecol. 27: 107-119.

Desantis, L.M., Bowman, J., Lahoda, C.V., Boonstra, R., and Burness, G. 2016. Responses of

New World flying squirrels to the acute stress of capture and handling. J. Mammal. 97:

80-88.

Page 17 of 26



Draft

18

Hammond, G.L. 2016. Plasma steroid-binding proteins: primary gatekeepers of steroid hormone

action. J. Endocrinol. 230: R13-R25.

Hammond, G.L., Smith, C.L., Lähteenmäki, P.L.A., Grolla, A., Warmels-Rodenhiser, S.,

Hodgert, H., Murai, J.T., and Siiteri, P.K. 1994. Squirrel monkey corticosteroid-binding

globulin: primary structure and comparison with the human protein. Endocrinology, 134:

891–898.

Kerhoulas, N.J., and Arbogast, B.S. 2010. Molecular systematics and Pleistocene biogeography

of Mesoamerican flying squirrels. J. Mammal. 91: 654-667.

Klosterman, L.L., Murai, J.T., and Siiteri, P.K. 1986. Cortisol levels, binding, and properties of

corticosteroid-binding globulin in the serum of primates. Endocrinology, 118: 424–434.

Lin, H., Muller, Y.A., and Hammond, G.L. 2010. Molecular and structural basis of steroid

hormone binding and release from corticosteroid-binding globulin. Mol. Cell.

Endocrinol. 316: 3–12.

Mendel, C.M. 1992. The free hormone hypothesis. Distinction from the free hormone transport

hypothesis. J. Androl. 13: 107–116.

Page 18 of 26



Draft

19

Mercer, J.M., and Roth, V.L. 2003. The effects of Cenozoic global change on squirrel

phylogeny. Science, 299: 1568-1572.

Moisan, M.P. 2010. Genotype-phenotype associations in understanding the role of

corticosteroid-binding globulin in health and disease animal models. Mol. Cell.

Endocrinol. 316: 35-41.

Pemberton, P.A., Stein, P.E., Pepys, M.B., Potter, J.M., and Carrell, R.W. 1988. Hormone

binding globulins undergo serpin conformational change in inflammation. Nature, 336:

257–258.

Perogamvros, I., Kayahara, M., Trainer, P.J., and Ray, D.W. 2011. Serum regulates cortisol

bioactivity by corticosteroid-binding globulin-dependent and independent mechanisms,

as revealed by combined bioassay and physicochemical assay approaches. Clin.


Perogamvros, I., Ray, D.W., and Trainer, P.J. 2012. Regulation of cortisol bioavailability –

effects on hormone measurement and action. Nat. Rev. Endocrinol. 8: 717-727.

Rasband, W.S. 2015. ImageJ version 1.49. U.S. National Institutes of Health, Bethesda,

Maryland, USA. http://imagej.nih.gov/ij/, 1997-2015.

Robinson, P.A., Hawkey, C., and Hammond, G.L. 1985. A phylogenetic study of the structural

Page 19 of 26



Draft

20

and functional characteristics of corticosteroid binding globulin in primates. J.


Rosner, W., Pugeat, M.M., Chrousos, G.P., and Khan, M.S. 1986. Steroid-binding proteins in

primate plasma. Endocrinology, 118: 513–517.

Sapolsky, R.M., Romero, L.M., and Munck, A.U. 2000. How do glucocorticoids influence stress

responses? Integrating permissive, suppressive, stimulatory, and preparative actions.

Endocr. Rev. 21: 55–89.

Scammell, J.G., Denny, W.B., Valentine, D.L., and Smith, D.F. 2001. Overexpression of the

FK506-binding immunophilin FKBP51 is the common cause of glucocorticoid resistance

in three new world primates. Gen. Comp. Endocrinol. 124: 152–165.

Smith, C.L., and Hammond, G.L. 1992. Hormonal regulation of corticosteroid-binding globulin

biosynthesis in the male rat. Endocrinology, 130: 2245-2251.

Westphal, U. 1983. Corticosteroid-binding globulin. Mol. Cell. Biochem. 55: 145–157.

Page 20 of 26



Draft

21

Fig. 1 Phylogenetic relationship among the five Sciurid species used in this study. Flying

squirrels are a monophyletic group most closely related to North American tree squirrels.

Redrawn from Kerhoulas and Arbogast (2010), and Mercer and Roth (2003). Branch lengths and

further details concerning clades and species within the family Sciuridae can be found in the

given references.

Fig. 2 Relative plasma CBG levels via western blotting (left y-axis), and plasma total cortisol

concentrations for the same individuals (right y-axis) for the five study species: southern flying

squirrels (Glaucomys volans; Southern), northern flying squirrels (Glaucomys sabrinus;

Northern), red squirrels (Tamiasciurus hudsonicus; Red), eastern grey squirrels (Sciurus

carolinensis; Grey), and eastern chipmunks (Tamias striatus; Chipmunk). Data are means ±

SEM. Invisible error bars are hidden behind the symbol for the mean. Uppercase letters denote

significant differences in total cortisol concentrations, and lowercase letters denote significant

differences among species in CBG values.

Fig. 3 Representative western blot of plasma CBG levels in adult individuals of the five study

species in the non-breeding season: southern flying squirrels (Glaucomys volans; Southern),

northern flying squirrels (Glaucomys sabrinus; Northern), red squirrels (Tamiasciurus

hudsonicus; Red), eastern grey squirrels (Sciurus carolinensis; Grey), and eastern chipmunks

(Tamias striatus; Chipmunk). All are males except for the grey squirrel on the left, which is a

female.

Page 21 of 26



Draft

22

Fig. 4 Relationship between relative plasma CBG levels and total cortisol concentrations for

each of the five study species: southern flying squirrels (Glaucomys volans; Southern), northern

flying squirrels (Glaucomys sabrinus; Northern), red squirrels (Tamiasciurus hudsonicus; Red),

eastern grey squirrels (Sciurus carolinensis; Grey), and eastern chipmunks (Tamias striatus;

Chipmunk). The solid line shows the linear regression for the three comparative species (Red,

Grey, Chipmunk). The dotted line is an extrapolation of the regression line, shown for ease of

comparison to data from the two flying squirrel species.

Page 22 of 26



Draft

Fig. 1 Phylogenetic relationship among the five Sciurid species used in this study. Flying squirrels are a monophyletic group most closely related to North American tree squirrels. Redrawn from Kerhoulas and Arbogast (2010), and Mercer and Roth (2003). Branch lengths and further details concerning clades and

species within the family Sciuridae can be found in the given references.

541x406mm (72 x 72 DPI)

Page 23 of 26



Draft

Fig. 2 Relative plasma CBG levels via western blotting (left y-axis), and plasma total cortisol concentrations for the same individuals (right y-axis) for the five study species: southern flying squirrels (Glaucomys

volans; Southern), northern flying squirrels (Glaucomys sabrinus; Northern), red squirrels (Tamiasciurus

hudsonicus; Red), eastern grey squirrels (Sciurus carolinensis; Grey), and eastern chipmunks (Tamias striatus; Chipmunk). Data are means ± SEM. Invisible error bars are hidden behind the symbol for the

mean. Uppercase letters denote significant differences in total cortisol concentrations, and lowercase letters denote significant differences among species in CBG values.

146x85mm (300 x 300 DPI)

Page 24 of 26



Draft

Fig. 3 Representative western blot of plasma CBG levels in adult individuals of the five study species in the non-breeding season: southern flying squirrels (Glaucomys volans; Southern), northern flying squirrels (Glaucomys sabrinus; Northern), red squirrels (Tamiasciurus hudsonicus; Red), eastern grey squirrels

(Sciurus carolinensis; Grey), and eastern chipmunks (Tamias striatus; Chipmunk). All are males except for the grey squirrel on the left, which is a female.

541x406mm (72 x 72 DPI)

Page 25 of 26



Draft

Fig. 4 Relationship between relative plasma CBG levels and total cortisol concentrations for each of the five study species: southern flying squirrels (Glaucomys volans; Southern), northern flying squirrels (Glaucomys

sabrinus; Northern), red squirrels (Tamiasciurus hudsonicus; Red), eastern grey squirrels (Sciurus

carolinensis; Grey), and eastern chipmunks (Tamias striatus; Chipmunk). The solid line shows the linear regression for the three comparative species (Red, Grey, Chipmunk). The dotted line is an extrapolation of

the regression line, shown for ease of comparison to data from the two flying squirrel species.

158x109mm (300 x 300 DPI)

Page 26 of 26



Documents

DraftDraft 1 Corticosteroid-binding globulin levels in North American sciurids: implications for the flying squirrel stress axis Lanna M. Desantis1*, Jeff Bowman2, Erin Faught3, Rudy